\documentclass[12pt]{article}
%===============================================================================
%=== (1) Packages
%===============================================================================
\usepackage[english]{babel}
\selectlanguage{english}

\usepackage{amsfonts}
\usepackage{amsmath}
\usepackage{appendix}
\usepackage{bm}
\usepackage{booktabs}
\usepackage[usenames, dvipsnames]{color}
\usepackage{graphicx}
\usepackage[margin=0.8in]{geometry}
\usepackage{epstopdf}
\epstopdfsetup{update}
\usepackage{helvet}
\usepackage{indentfirst}
\usepackage{pdflscape}
\usepackage{morefloats}
\usepackage{natbib} \bibliographystyle{aea}
\usepackage{setspace}
\usepackage{subcaption}
\usepackage[capposition=top]{floatrow}
\usepackage{subfloat}
\usepackage{pdfpages}
\usepackage{multirow}
\usepackage[bottom]{footmisc}
\usepackage[table]{xcolor}

\usepackage{pgfplotstable}
\usepackage[hypertexnames=false]{hyperref}
\usepackage[normalem]{ulem}

%===============================================================================
%=== (2) Specific Instructions
%===============================================================================
\hypersetup{
    colorlinks=true,      
    linkcolor=BlueViolet, 
    citecolor=BlueViolet, 
    filecolor=BlueViolet, 
    urlcolor=BlueViolet   
}


\newcommand{\appendixpagenumbering}{
  \break
  \pagenumbering{arabic}
  \renewcommand{\thepage}{\thesection\arabic{page}}
}
\setlength\parindent{0.15in}
\setlength\parskip{0.12in}


\usepackage{fontspec}
\setmainfont{Times New Roman}

\begin{document}


\begin{figure}[ht!]
  \begin{center}
    \centering
    \caption{Usage of Gestational Component of ChCC by Month}
    \includegraphics[scale=0.9]{../replication/results/main/ChCCtime.eps}
    \label{fig:coverage}
  \end{center}
  \vspace{-5mm}
  \floatfoot{\textsc{Notes to Figure \ref{fig:coverage}}: Program usage
    by month and municipality, and proportion of all births covered nation-wide
    is calculated from administrative data provided by the Ministry of Social
    Development.  This captures the proportion of all mothers giving birth each
    month who participated in the pre-natal components of ChCC prior to giving
    birth.  The program did not exist prior to 2007.  Additional details can be
    found in section \ref{scn:data} of this paper.  Geographic distribution of
    municipal roll-out is provided in Appendix Figure \ref{fig:map}.
  }
\end{figure}

\input{../replication/results/main/rollout_analysis.tex}

\begin{table}[h]
  \begin{center}
    \caption{Summary Statistics: Birth and Chile Crece Contigo Data}
    \label{tab:sumstats}
    \begin{tabular}{lccccc} \toprule
      & N& Mean & Std. Dev. & Min & Max \\ \midrule
      \input{../replication/results/main/SummaryMunicipal.tex} \midrule
      \multicolumn{6}{p{14.8cm}}{{\footnotesize \textsc{Notes to Table
            \ref{tab:sumstats}}: Summary Statistics are displayed for
          municipality by month averages for
          each month from January 2003 to December 2010.  Averages are
          displayed for each municipality in which there is at least one
          birth in the given month.  The average number of births by
          municipality and month is displayed above.  There are 346 municipalities
          in Chile, and hence a maximum number of observations of 346
          municipalities $\times$ 8 years $\times$ 12 months, or 33,216
          municipality $\times$ month observations.  The difference between
          this maximum and the observed number of observations are cases
          where no births occurred.  Uncollapsed micro-data on births
          consists of 1,917,085 observations between 2003 and 2010.
          Additional details on this birth data is provided in
          Appendix \ref{app:context}.  Proportion enrolled in ChCC
          refers to the average proportion of births in each municipality
          which were covered by ChCC \emph{in utero} during the entire
          period of 2003-2010, and so is always zero prior to the implementation
          of ChCC in 2007/2008.
      }} \\ \bottomrule
      \end{tabular}
  \end{center}
\end{table}

\input{../replication/results/main/comunaDD.tex}

\begin{landscape}
\begin{table}[h]
  \begin{center}
    \caption{Alternative Specifications: Diff-in-diff Estimates of Program Impacts}
    \label{tab:AltSpecs}
    \scalebox{0.95}{
    \begin{tabular}{lccccccccc} \toprule
      &(1)&(2)&(3)&(4)&(5)&(6)&(7)&(8)&(9)\\ \midrule
      \multicolumn{10}{l}{\textbf{Panel A: Birth Weight}} \\
      \input{../replication/results/main/Alt_peso.tex}
      \multicolumn{10}{l}{\textbf{Panel B: LBW}} \\
      \input{../replication/results/main/Alt_lbw.tex}
      \multicolumn{10}{l}{\textbf{Panel C: Size}} \\
      \input{../replication/results/main/Alt_talla.tex}
      \multicolumn{10}{l}{\textbf{Panel D: Gestation}} \\
      \input{../replication/results/main/Alt_gestation.tex}
      \multicolumn{10}{l}{\textbf{Panel E: Premature}} \\
      \input{../replication/results/main/Alt_premature.tex}
      \multicolumn{10}{l}{\textbf{Panel F: Infant Mortality}} \\
      \input{../replication/results/main/Alt_fDeathRate.tex}
      \midrule
      Municipal and Time FEs      & Y & Y & Y & Y & Y & Y & Y & Y & Y \\
      Time-Varying Controls       &   & Y &   &   & Y &   &   &   &   \\  
      Region Time Trends          &   &   & Y &   &   &   &   &   &   \\
      Region $\times$ Year FEs    &   &   &   & Y & Y &   &   &   &   \\
      Municipal Linear Trends     &   &   &   &   &   & Y &   &   &   \\
      Municipal Quadratic Trends  &   &   &   &   &   &   & Y &   &   \\
      Municipal $\times$ Year FEs &   &   &   &   &   &   &   & Y &   \\
      Weighting by Pregnancies    &   &   &   &   &   &   &   &   & Y \\
      \bottomrule
      \multicolumn{10}{p{21.8cm}}{{\footnotesize \textsc{Notes to Table \ref{tab:AltSpecs}}:
          Each specification is estimated by DD using
          municipal-level averages by month, and weights for the number of observations
          in each cell.  Column 1 replicates results from Table \ref{mDD}, and then
          columns 2-8 include additional controls, time trends, or fixed effects.
          Column 9 weights by the number of pregnancies, rather than births.
          Regions in Chile are the second-level administrative district, of which there are
          15.  Municipalities are within districts (analogous to states and counties
          in other countries), and there are 346 municipalities in Chile.  In each
          case where time trends are included, these are split for pre- and post-reform
          periods.  The
          most demanding specification allows for a separate fixed effect for each
          municipality in each year under study, given that there are twelve
          observations for each municipality in each year.  Time-varying controls are
          collected from the Government of Chile's National System for Municipal Information,
          and are available for each municipality in each year.  These controls consist
          of total transfers for education and health, the proportion of each municipality
          enrolled in the public health system (FONASA), the proportion enrolled in school,
          a pre-determined poverty index calculated by the government, and the coverage
          of drinking water.  Standard errors are clustered by Municipality. Refer
          to Table \ref{mDD} for additional notes.}} \\
    \end{tabular}}
  \end{center}
\end{table}
\end{landscape}

\begin{figure}[htpb!]
  \begin{center}
    \caption{Placebo Tests}
    \label{placebo}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/main/placebolag_peso.eps}
      \caption{Birth Weight}
      \label{placebo-peso}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/main/placebolag_lbw.eps}
      \caption{LBW}
      \label{placebo-lbw}
    \end{subfigure}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/main/placebolag_gestation.eps}
      \caption{Gestation}
      \label{placebo-gest}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/main/placebolag_fDeathRate.eps}
      \caption{Fetal Deaths}
      \label{placebo-fdeaths}
    \end{subfigure}
  \end{center}
  \vspace{-6mm}
  \floatfoot{\textsc{Notes to Figure \ref{placebo}}: Each point estimate
    and resulting confidence interval display the impact of a placebo test
    where the share of Chile Crece Contigo enrollees is lagged
    $j\in \{1,\ldots,40\}$ months, where $j$ is displayed on the horizontal
    axis. Each placebo test is estimated following the principal
    specification displayed in Table \ref{mDD}.  Additional notes relating
    to each model can be found in Table \ref{mDD}.}
\end{figure}


\begin{table}[ht!]
  \begin{center}
    \caption{Impacts by Vulnerability Quintile}
    \label{tab:FPS}
    \begin{tabular}{lccccc} \toprule
      &(1)&(2)&(3)&(4)&(5)\\
      & Weight &\ \ LBW \ \ &\ \  Size \  \ & Gestation & Premature \\ \midrule
      \multicolumn{6}{l}{\textbf{Panel A: Quintile 1 (20\% Most Vulnerable)}} \\
      \input{../replication/results/main/FPS_1.tex}
      \\
      \multicolumn{6}{l}{\textbf{Panel B: Quintiles 1-2 (40\% Most Vulnerable)}} \\
      \input{../replication/results/main/FPS_2.tex}
      \\
      \multicolumn{6}{l}{\textbf{Panel C: Quintiles 1-3 (60\% Most Vulnerable)}} \\
      \input{../replication/results/main/FPS_3.tex}
      \\
      \multicolumn{6}{l}{\textbf{Panel D: Quintile 4+ (Non-targeted)}} \\
      \input{../replication/results/main/FPS_4.tex}
      \bottomrule
      \multicolumn{6}{p{14.7cm}}{{\footnotesize \textsc{Notes to Table \ref{tab:FPS}}:
          Identical specifications are estimated as in Table \ref{mDD}, however now each
          model is estimated using \emph{only} observations which meet the criteria
          defined in panel headings. Classification of the 20\%, 40\%, and 60\% most vulnerable
          is based on the Government of Chile's official scoring based on the
          \emph{Ficha de Protecci\'on Social} (FPS, or Social Protection Score in English),
          which is used to classify the degree of benefits received by families in ChCC.
          The official cut-off for the 20\% most vulnerable is a score of 8,500 points or lower
          on the social protection score, and for the 40\% and 60\% most vulnerable is a
          score of 11,734 or 13,484 points or lower (respectively).  Any mother with a score
          above 13,484 (or who has not
          applied for a score) is not in the targeted group. Additional discussion of the
          FPS is available in \citet{Herreraetal2010}.}} \\
    \end{tabular}
  \end{center}
\end{table}


\begin{figure}[htpb!]
  \begin{center}
    \caption{Regression Discontinuity Plots at Vulnerability Score Cut-off}
    \label{fig:RD}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.89]{../replication/results/main/cutoffPost_peso.eps}
      \caption{Birth Weight}
      \label{rd-peso}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.89]{../replication/results/main/cutoffPost_lbw.eps}
      \caption{LBW}
      \label{rd-lbw}
    \end{subfigure}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.89]{../replication/results/main/cutoffPost_semanas.eps}
      \caption{Gestation}
      \label{rd-gest}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.89]{../replication/results/main/cutoffPost_talla.eps}
      \caption{Size}
      \label{rd-size}
    \end{subfigure}
  \end{center}
  \floatfoot{\textsc{Notes to Figure \ref{fig:RD}}: Plots documented average health at
    birth based on the binned Social Protection Score of mothers.  The vertical
    dashed line is drawn at 13,484 points, the cut-off for Chile Crece Contigo
    preferential services.  Circles represent raw averages in bins (bins of 55 points are
    used), and solid lines represent a polynomial fit of these binned points.  Formal
    tests of regression discontinuity \ref{tab:RDestimates} models are provided
    in Table \ref{tab:RDestimates}.}
\end{figure}


\begin{table}[htpb!]
  \centering
  \caption{Regression Discontinuity Estimates of Preferential ChCC Cut-off}
  \label{tab:RDestimates}
  \begin{tabular}{lccccc}
    \toprule
    &Weight & LBW & Size & Gestation & Prematurity \\
    &(1)&(2)&(3)&(4)&(5)\\ \midrule
    \multicolumn{6}{l}{\textbf{Panel A: Quadratic Polynomial in Running Variable}} \\
    \input{../replication/results/main/RDestimatesQuad.tex} \\
    \multicolumn{6}{l}{\textbf{Panel B: Local Linear with CCT Optimal Bandwidth}}\\
    \input{../replication/results/main/RDestimatesLinear.tex} \midrule
    \multicolumn{6}{p{15cm}}{{\footnotesize Notes: Panel A displays regression
        discontinuity estimates based on intensive margin program participation 
        using a global polynomial estimate with a quadratic fit on either side of
        cut-off to capture evolution of the running-variable (quadratic is allowed
        to vary on either side).  Panel B displays local linear estimates based on
        \citet{Calonicoetal2014}.  The optimal bandwidth is
        displayed at the foot of panel B, along with the number of observations
        located within this bandwidth of the cut-off.  All estimates are based on
        the Social Protection Score cut-off point of 11,384 points.}}
  \end{tabular}
\end{table}

  
\begin{figure}[htpb!]
  \begin{center}
    \caption{Policy Impact Across the Health Distribution}
    \label{quintiles}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/main/Birthweight_Cutoffs.eps}
      \caption{Birth Weight}
      \label{quintiles-level}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/main/Gestation_Cutoffs.eps}
      \caption{Gestation}
      \label{quintiles-log}
    \end{subfigure}
  \end{center}
  \floatfoot{\textsc{Notes to Figure \ref{quintiles}}: Point estimates and 95\%
    confidence intervals are presented of the impact of Chile Crece Contigo on
    birth weight and gestational length at different points of the distribution.
    Each specification follows equation \ref{eqn:DD}, however instead of using
    mean birth weight or gestational length in each municipality, uses the
    proportion of births exceeding determined cut-points of the distribution
    (displayed on the horizontal axis) as the dependent variable of interest.
    Panel \ref{quintiles-level} displays the estimates when considering birth
    weight, while panel \ref{quintiles-log} presents estimate for gestational
    length.  For additional details, refer to notes to Table \ref{mDD}.
  }
\end{figure}


\begin{landscape}
\begin{table}[h]
  \begin{center}
    \caption{Scarring versus Selection: Simulating Unselected Birth Outcomes}
    \label{tab:alderman}
    \scalebox{0.95}{
    \begin{tabular}{lccccccccc} \toprule
      &5\%&10\%&25\%&50\%&60\%&70\%&80\%&90\%&100\% \\
      &(1)&(2)&(3)&(4)&(5)&(6)&(7)&(8)&(9)\\ \midrule
      \multicolumn{10}{l}{\textbf{Panel A: Low Birth Weight}} \\
      \input{../replication/results/main/selection_LBW.tex}
      \multicolumn{10}{l}{\textbf{Panel B: Prematurity}} \\
      \input{../replication/results/main/selection_prem.tex}
      \bottomrule
      \multicolumn{10}{p{20.6cm}}{{\footnotesize \textsc{Notes to Table \ref{tab:alderman}}:
          Each regression uses the full sample of birth and fetal death data, however removes
          a portion of births in the post-ChCC period assumed to be `selectively surviving'
          due to ChCC.  In each column it is assumed that $x$\% of these selectively
          surviving births would have been of low birth weight (panel A) or born prematurely
          (panel B).  The percentage assumed to meet this condition is indicated in column
          headers.  In both cases the outcome variable (proportion of low birth weight and
          proportion of premature births) is multiplied by 100 for ease of visualisation.
          Means of dependent variables under each assumed counterfactual are indicated at
          the foot of each panel. All other details follow those in Table \ref{mDD}.}} \\
    \end{tabular}}
  \end{center}
\end{table}
\end{landscape}



\begin{table}
  \caption{Costs and Estimated Impacts of Selected Early-Life Programs}
  \label{tab:impactsOther}
  \begin{tabular}{lcccc} \toprule
    Reference & Estimated & Cost per    & Estimated      \\
    & Impact    & Participant & Cost per gram  \\ \midrule
    \multicolumn{5}{l}{\textbf{Supplemental Nutrition Program for Women, Infants and Children (WIC, US)}} \\
    Rossin-Slater (2013) & 27.30 (7.98)               & \$405 USD & \$14.8 \\
    Hoynes et al.\ (2011)   & 28.75 (15.13)              & \$405 USD & \$14.1 \\
    &&&\\
    \multicolumn{5}{l}{\textbf{PANES (Uruguay)}} \\
    Amarante et al.\ (2016) & 30.83 (18.44)  & \$918 USD & \$29.8 \\
    &&&\\
    \multicolumn{5}{l}{\textbf{Supplemental Nutrition Assistance Program (FSP, US)}} \\
    Almond et al.\ (2011)   & 8.96 (5.05) & \$1125 USD & \$125.6 \\
                             & 20.27 (6.89) & \$1125 USD & \$55.5 \\
    &&&\\
    \multicolumn{5}{l}{\textbf{Earned Income Tax Credit (EITC, US)}} \\
    Strully et al.\ (2010) & 15.70 (1.211)& \$1558 USD &  \$99.2\\
    Hoynes et al.\ (2015) & 9.95  (2.05)  & \$1558 USD & \$156.6  \\
    &&&\\    
    \multicolumn{5}{l}{\textbf{Chile Crece Contigo (Chile)}} \\
    Our estimates            & 10.09 (3.37)   & \$111 USD & \$11.0\\ \bottomrule
    \multicolumn{5}{p{15.4cm}}{{\footnotesize\textsc{Notes}: Estimates from
        Hoynes et al. (2015) refer to single women with no more than a
        high-school education (the ``high impact'' group, with highest
        eligibility for policy use). Two estimates are presented for
        Almond et al.\ (2011), given that their results are presented by
        race.  The top line refers only to black mothers, while the bottom
        line refers only to white mothers.  Estimates for black mothers are
        based on the most recent estimates presented by the authors in their
        Erratum.  All US program costs are expressed in US dollars, and
        non-US program costs (Chile and Uruguay) are denoted in PPP adjusted
        US dollars.  PPP adjusted costs are higher than non-PPP adjusted
        costs, so this results in a conservative estimates of costs per gram.
        Similar estimates and additional calculation details are presented in
        Clarke et al. (2017) for the WIC and FSP only.}}\\
  \end{tabular}
\end{table}

\begin{landscape}
\input{../replication/results/main/ChCC_Mechanism.tex}
\end{landscape}


\clearpage
\setcounter{table}{0}
\renewcommand{\thetable}{A\arabic{table}}
\setcounter{figure}{0}
\renewcommand{\thefigure}{A\arabic{figure}}
\pagenumbering{arabic}

\renewcommand{\thepage}{A\arabic{page}}
\appendix
\section*{Online Appendices}

\section{Appendix Tables and Figures}

\input{../replication/results/appendix/FONASAtest.tex}

\begin{landscape}
\input{../replication/results/appendix/comunaDD-together.tex}
\end{landscape}

%%XXX
\begin{table}[htpb!]
  \begin{center}
    \caption{Summary Statistics by Trimester: Birth and Chile Crece Contigo Data}
    \label{tab:sumstatsTri}
    \begin{tabular}{lccccc} \toprule
      & N& Mean & Std. Dev. & Min & Max \\ \midrule
      \input{../replication/results/appendix/SummaryMunicipal-trimester.tex} \bottomrule
      \multicolumn{6}{p{14.4cm}}{{\footnotesize \textsc{Notes to Table
            \ref{tab:sumstatsTri}}: Summary Statistics are displayed for
          municipality by trimesterly averages for
          each trimester from January 2003 to December 2010.  Trimesters
          refer to January-March, April-June, July-September, and
          October-December.  For additional notes, refer to Table
          \ref{tab:sumstats} which provides summary statistics at the
          municipality by month level.
      }}
      \end{tabular}
  \end{center}
\end{table}

\begin{landscape}
\input{../replication/results/appendix/comunaDD-trimester.tex}
\end{landscape}

\begin{landscape}
\input{../replication/results/appendix/comunaDD-short.tex}
\end{landscape}

\begin{landscape}
\input{../replication/results/appendix/comunaDD-shortIV.tex}
\end{landscape}

\begin{landscape}
\begin{table}[h]
  \begin{center}
    \caption{Examining Robustness of Impacts on Birth weight to removal of extreme values}
    \label{tab:extremeRobust}
    \scalebox{0.95}{
    \begin{tabular}{lccccccccc} \toprule
      &(1)&(2)&(3)&(4)&(5)&(6)&(7)&(8)&(9)\\ \midrule
      \multicolumn{10}{l}{\textbf{Panel A: Winsorizing at 1\textsuperscript{st} and 99\textsuperscript{th} Percentiles}} \\
      \input{../replication/results/main/Alt_peso_w.tex}
      \multicolumn{10}{l}{\textbf{Panel B: Trimming at 1\textsuperscript{st} and 99\textsuperscript{th} Percentiles}} \\
      \input{../replication/results/main/Alt_peso_trim.tex}
      \midrule
      Municipal and Time FEs      & Y & Y & Y & Y & Y & Y & Y & Y & Y \\
      Time-Varying Controls       &   & Y &   &   & Y &   &   &   &   \\  
      Region Time Trends          &   &   & Y &   &   &   &   &   &   \\
      Region $\times$ Year FEs    &   &   &   & Y & Y &   &   &   &   \\
      Municipal Linear Trends     &   &   &   &   &   & Y &   &   &   \\
      Municipal Quadratic Trends  &   &   &   &   &   &   & Y &   &   \\
      Municipal $\times$ Year FEs &   &   &   &   &   &   &   & Y &   \\
      Weighting by Pregnancies    &   &   &   &   &   &   &   &   & Y \\
      \bottomrule
      \multicolumn{10}{p{21.2cm}}{{\footnotesize \textsc{Notes to Table \ref{tab:extremeRobust}}:
          Each specification follows models documented in Panel A of Table \ref{tab:AltSpecs},
          however here examining robustness of the birth weight results to outliers.  In panel A,
          average birth weight in each municipality (the outcome of interest) is Winsorized at the
          1\textsuperscript{st} and 99\textsuperscript{th} percentiles implying that observations
          more extreme than these values are replaced with the values of these percentiles.  In
          this case the full sample of 31,805 observations is used.  In panel B, the sample is
          trimmed at the 1\textsuperscript{st} and 99\textsuperscript{th} percentiles, and so
          observations more extreme than these values are simply removed from the sample.  In
          this case, the estimation sample consists of 31,169 municipality $\times$ year cells.
          In both specifications, average municipal birth weight ranges from a minimum of 2,844
          grams, to a maximum of 3,825 grams.
          Refer to Table \ref{tab:AltSpecs} for additional notes.}} \\
    \end{tabular}}
  \end{center}
\end{table}
\end{landscape}

%%XXX
\begin{table}[htpb!]
  \caption{Adjusting For Multiple Hypothesis Testing}
  \label{tab:MultHyp}
  \begin{center}
      \begin{tabular}{lcccccc} \toprule
        &  Index & \multicolumn{5}{c}{Original Variables} \\ \cmidrule(r){2-7}
        &Anderson & Birth & LBW & Birth & Weeks & Premature \\
        &Index    & Weight&     & Size  & Gestation & \\ \midrule
        \multicolumn{7}{l}{\textsc{Panel A: Municipal-Level Analysis}} \\
        $p$-value  (Original)    & \input{../replication/results/appendix/MC_DD_porig.tex}
        $p$-value  (Corrected) & \input{../replication/results/appendix/MC_DD_pRW.tex}
        \multicolumn{7}{l}{\textsc{Panel B: Individual-Level Analysis}} \\
        $p$-value  (Original)    & \input{../replication/results/appendix/MC_FE_porig.tex}
        $p$-value  (Corrected) & \input{../replication/results/appendix/MC_FE_pRW.tex}
        \midrule
        \multicolumn{7}{p{14.6cm}}{{\footnotesize \textsc{Notes}: Corrected $p$-values based
            on original variables are calculated using the \citet{RomanoWolf2005} technique to
            control the Family Wise Error Rate of hypothesis tests, implemented by \citet{Clarke2016}.
            The \citet{Anderson2008}
            index converts the multiple dependent variables into a single dependent variable
            (index) giving more weight to variables which provide more independent variation.
            The specification of each regression follows Table \ref{mDD} (panel A), and
            Appendix Table \ref{mFE} (panel B).}}
        \\ \bottomrule
    \end{tabular}
  \end{center}
\end{table}


\begin{landscape}
\input{../replication/results/appendix/comunaDD-Available.tex}
\end{landscape}

\begin{landscape}
\begin{table}
  \begin{center}
    \caption{IV Estimates Using Lagged ChCC Enrollment}
    \label{tab:ChCC_IV}
    \begin{tabular}{lcccccc} \toprule 
      & (1)    & (2) & (3)  & (4)       & (5)       & (6)         \\
      & Weight & LBW & Size & Gestation & Premature & Fetal Death \\ \midrule
      \multicolumn{7}{l}{\textbf{Second Stage Estimates}} \\
      \input{../replication/results/appendix/comunaDD-IV.tex} \\
      \multicolumn{7}{l}{\textbf{First Stage Estimates}} \\
      \input{../replication/results/appendix/comunaDD-first.tex}
      \midrule
      \multicolumn{7}{p{17.6cm}}{{\footnotesize \textsc{Notes}:
          Difference-in-difference estimates are presented following the
          results of Table \ref{mDD}.  However, here the Proportion of ChCC
          Coverage among births in a given month and municipality is
          instrumented with lagged ChCC coverage from the same municipality.
          The 2SLS results along with standard errors clustered by municipality
          are displayed in the top panel of the Table.  The second panel
          documents the first stage results of regression ChCC coverage
          on its lagged value.  The associated first stage F-statistic and its
          p-value are documented at the foot of the table.
      }} \\ \bottomrule
    \end{tabular}
  \end{center}
\end{table}
\end{landscape}


\clearpage
\begin{table}[htpb!]
  \caption{Correction for Multiple Hypothesis Testing in Distributional Estimates}
  \label{tab:RWdist}
\begin{tabular}{lcclcc} \toprule
  \multicolumn{3}{c}{Birth Weight}&\multicolumn{3}{c}{Gestation} \\ \cmidrule(r){1-3} \cmidrule(r){4-6}
  Cut-off & Original & Romano Wolf &   Cut-off & Original & Romano Wolf \\
  & $p$-value & $p$-value  &           & $p$-value& $p$-value \\ \midrule
  1000 &0.4592 & 0.6573 & 30 & 0.6905 & 0.7922 \\
  1250 &0.5786 & 0.7493 & 31 & 0.6245 & 0.7822 \\
  1500 &0.7191 & 0.8492 & 32 & 0.3666 & 0.5315 \\
  1750 &0.0632 & 0.0639 & 33 & 0.0464 & 0.0370 \\
  2000 &0.0014 & 0.0000 & 34 & 0.1695 & 0.2398 \\
  2250 &0.0135 & 0.0060 & 35 & 0.0804 & 0.0739 \\
  2500 &0.0737 & 0.0759 & 36 & 0.0539 & 0.0410 \\
  2750 &0.2736 & 0.4116 & 37 & 0.2337 & 0.3417 \\
  3000 &0.1169 & 0.1299 & 38 & 0.2651 & 0.3596 \\
  3250 &0.2212 & 0.3487 & 39 & 0.0477 & 0.0370 \\
  3500 &0.0056 & 0.0010 & 40 & 0.0005 & 0.0000 \\
  3750 &0.0030 & 0.0000 & 41 & 0.5312 & 0.7493 \\
  4000 &0.0221 & 0.0120 & 42 & 0.9967 & 0.9960 \\
  4250 &0.0167 & 0.0070 &    && \\
  4500 &0.0144 & 0.0060 &    && \\
  4750 &0.9501 & 0.9281 &    && \\
  5000 &0.4313 & 0.6573 &    && \\ \midrule
  \multicolumn{6}{p{12.6cm}}{{\footnotesize \textsc{Notes to Table \ref{tab:RWdist}}:
      Un-adjusted and multiple-hypothesis test adjusted $p$-values are displayed
      corresponding to the estimates and standard errors displayed in Figure
      \ref{quintiles}.  Unadjusted $p$-values refer to the $p$-value on ChCC in
      each regression where the outcome variable is birth weight or gestation
      exceeding the listed cut-off.  Romano Wolf adjusted $p$-values are based on
      a null re-sampled distribution as described in \citet{RomanoWolf2005}.  We
      re-sample using 1000 bootstrap samples.}} \\ \bottomrule
\end{tabular}
\end{table}


\begin{table}[htpb!]
  \caption{Costs of ChCC Per Participant in Gestational Program}
  \label{tab:spending}
  \scalebox{0.94}{
  \begin{tabular}{lcccc} \toprule
    &2007&2008&2009&2010 \\ \midrule
    \multicolumn{5}{l}{\textit{Panel A: All Amounts in 1000s of Chilean Pesos}} \\
    Costs Associated with PADBP&1,969,162&6,116,663&14,231,107&14,444,574\\
    Costs Ministry of Planning &1,001,810&2,529,976&2,604,131&4,197,607\\
    Massive Education Program &20,000&195,640&261,462&196,624\\
    \textbf{Total Prenatal Development Components} &2,990,972&8,842,279&17,096,700&18,838,805\\
    Total Budget (ChCC)&67,903,331&126,446,362&159,660,473&214,505,550\\
    Total Budget/1000 (All Chile)&17,883,154& 20,650,579 & 23,406,879& 25,651,970\\
    Total Women Participating during Gestation &47,683&166,900&171,811&171,799\\
    Proportion of all Participants in Pre-natal Care &1&0.449&0.307&0.303\\
    Cost per Pre-Natal Participant &62,726 & 24,714 & 30,549 & 33,116\\
    &&&&\\
    \multicolumn{5}{l}{\textit{Panel B: All Amounts in US Dollars}} \\
    Costs Associated with PADBP                    &3,702,025  &12,288,376 &22,257,451&28,470,255\\
    Costs Ministry of Planning                     &1,883,403  &5,082,722  &4,072,861 &8,273,483\\
    Massive Education Program                      &37,600     &393,041    &408,917   &387,546\\
    \textbf{Total Prenatal Development Components} &5,623,027  &17,764,139 &26,739,239&37,131,285\\
    Total Budget (ChCC)                            &127,658,262&254,030,741&249,708,980&422,790,439\\
    Total Budget/1000 (All Chile)                  &33,620,330 &41,487,013 & 36,608,359& 50,560,033\\
    Total Women Participating during Gestation     &47,683     &166,900&171,811&171,799\\
    Proportion of all Participants in Pre-natal Care &1&0.449&0.307&0.303\\
    Cost per Pre-Natal Participant &\$118 & \$50 & \$48 & \$65\\
    Cost per Pre-Natal Participant (PPP Adjusted) &\$192 & \$72 & \$87 & \$93\\
    \midrule
    \multicolumn{5}{p{18.1cm}}{{\footnotesize \textsc{Notes to Table \ref{tab:spending}:}
        Costs per pre-natal participant are calculated by dividing the pro-rata total costs
        of prenatal development components by the total number of participants in the pre-natal
        period.  Total prenatal development components are calculated as the sum of the costs
        of the PADBP program, fixed costs assigned to the Ministry of Planning, and the costs
        of the Massive Education program.  Costs are assigned pro-rata to pre-natal versus non
        pre-natal components using the proportion of all participants which are in the
        pre-natal
        period, rather than during years 1-5.  In the first year, the program only began in
        utero, so all costs are assigned to pre-natal development.  Budget details are all
        compiled from the ChCC final reports \citep{Arrietetal2013}, and historic budget
        laws (for example \citet{Presupuesto2007}).  Total participants during gestation as
        well as in the post-natal period are compiled from the Department of Health Statistics
        and Information from the Ministry of Health. PPP-adjusted costs are based on the
        World Bank's PPP conversion factor.}} \\ \bottomrule
  \end{tabular}}
\end{table}

\begin{landscape}
  \input{../replication/results/appendix/ChCC_Inputs.tex}
\end{landscape}

\input{../replication/results/appendix/GelbachMechanism.tex}


\clearpage
\thispagestyle{empty}
\begin{figure}[htpb!]
  \begin{center}
    \centering
    \caption{Program Roll-out by Date}
    \label{fig:map}
    \begin{subfigure}{.3\textwidth}
      \centering
      \includegraphics[scale=0.4]{../replication/results/appendix/Rollout_Time.eps}
      \caption{Full Country}
      \label{Chile}
    \end{subfigure}%
    \begin{subfigure}{.7\textwidth}
      \centering
      \includegraphics[scale=0.84]{../replication/results/appendix/Rollout_Time_RM.eps}
      \caption{Santiago Metropolitan Region}
      \label{RM}
    \end{subfigure}
  \end{center}
  \vspace{-5mm}
  \floatfoot{\textsc{Notes to Figure \ref{fig:map}}: Chile consists
    of 346 municipalities (``\emph{comunas}'') which are the lowest
    geographic administrative level with their own political administration.
    ChCC roll-out started in June 2007, and reached 159 of the 346
    municipalities in 2007 (chosen due to the availability of
    infrastructure) and then was rolled out to the remaining municipalities
    during 2008.  Precise roll-out dates are provided by the Ministry of Social
    Development of Chile.  The full country is displayed in the left-hand
    panel, and only the Metropolitan Region of Santiago (from the centre of
    the country) is displayed in the right-hand panel.}
\end{figure}

\begin{figure}[htpb!]
  \begin{center}
    \caption{ChCC Usage in Post-Implementation Period}
    \label{fig:usage}
    \includegraphics[scale=0.8]{../replication/results/appendix/chccUsage.eps}
  \end{center}
  \vspace{-6mm}
  \floatfoot{\textsc{Notes to Figure \ref{fig:usage}}: The density of ChCC usage
    by municipality over the entire post-treatment period is displayed.  Usage
    refers to the average proportion of all births in each municipality for which
    ChCC components 
    were accessed by the mother during the gestational period.  Usage data comes
    from The Ministry of Social Development's administrative data on public program
    use, and is averaged at the level of each municipality.  Refer to Figure
    \ref{ChCCenrol} for additional details regarding municipal level usage of
    ChCC components and municipal characteristics.
  }
\end{figure}

\begin{figure}[htpb!]
  \begin{center}
    \caption{Proportion of Births Attended in the Public Health System}
    \label{publicBirth}
    \includegraphics[scale=0.8]{../replication/results/appendix/birthsPublic.eps}
  \end{center}
  \vspace{-6mm}
  \floatfoot{\textsc{Notes to Figure \ref{publicBirth}}: Figures on
    the proportion of births in the public health system and all births
    nation-wide are provided monthly by the Department of Statistics and
    Health Information (DEIS) of the Ministry of Health of Chile. Monthly
    proportions are displayed for each month from January 2002 until
    December 2010.  The first vertical dotted line is the beginning of
    ChCC roll-out, while the second vertical dotted line is when ChCC
    reached the full country.
  }
\end{figure}


\begin{landscape}
\begin{figure}[htpb!]
  \begin{center}
    \caption{Municipal Characteristics and ChCC Enrollment}
    \label{ChCCenrol}
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_aguapotable.eps}
      \caption{Treated Piped Drinking Water}
      \label{agua}
    \end{subfigure}%
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_fonasaPC2064.eps}
      \caption{FONASA enrolments}
      \label{fonasa}
    \end{subfigure}%
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_FPSanioPC.eps}
      \caption{Proportion of FPS per Year}
      \label{FPS}
    \end{subfigure}
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_pobrezacasen.eps}
      \caption{Poverty}
      \label{pobreza}
    \end{subfigure}%
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_subEscPC.eps}
      \caption{Education Subvention}
      \label{subEsc}
    \end{subfigure}%
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_teen.eps}
      \caption{Proportion of Teen Births}
      \label{teen}
    \end{subfigure}
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_votop.eps}
      \caption{Vote Share (Mayor)}
      \label{votop}
    \end{subfigure}%
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_LCR.eps}
      \caption{Political Association}
      \label{LCR}
    \end{subfigure}%
    \begin{subfigure}{.33\textwidth}
      \centering
      \includegraphics[scale=0.44]{../replication/results/appendix/chcc_meduc.eps}
      \caption{Maternal Education}
      \label{teen}
    \end{subfigure}
  \end{center}
  \floatfoot{\textsc{Notes to Figure \ref{ChCCenrol}}: Each panel presents the proportion
    of Chile Crece Contigo enrollees in each municipality after the introduction of the
    program along with municipal level averages in a range of other social or political
    variables.  In each case, ChCC enrollment is displayed on the horizontal axis, and
    alternative outcomes on the vertical axis.
  }
\end{figure}
\end{landscape}

\begin{figure}[htpb!]
  \begin{center}
    \caption{Socioeconomic Quintiles and Health Distributions at Birth}
    \label{fig:healthSES}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.7]{../replication/results/appendix/weightdensity_Quintiles.eps}
      \caption{Birth weight}
      \label{FPSdens}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.7]{../replication/results/appendix/gestationdensity_Quintiles.eps}
      \caption{Gestational Period}
      \label{McCrary}
    \end{subfigure}
  \end{center}
  \vspace{-1cm}
    \floatfoot{\textsc{Notes to Figure \ref{fig:healthSES}}:
      Figures provide kernel density plots of birth weight (in grams) and weeks
      of gestation by quintiles of the Social Vulnerability Score.  Quintile 1 is
      the most vulnerable, and quintiles 4 and above are grouped into a single plot.
      Means for birth weight are 3350 grams, 3333 grams, 3317 grams and 3298 grams
      for quintiles 1, 2, 3 and 4+ respectively.  Similar means for gestational
      period are 38.66 weeks, 38.61 weeks, 38.55 weeks, and 38.43 weeks.
    }
\end{figure}


\begin{figure}[htpb!]
  \begin{center}
    \caption{Running Variable (FPS) in RDD}
    \label{RDDdensity}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.9]{../replication/results/appendix/FPS-hist.eps}
      \caption{Ficha de Protecci\'on Social: Density}
      \label{FPSdens}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.9]{../replication/results/appendix/McCraryTest.eps}
      \caption{McCrary Density Test}
      \label{McCrary}
    \end{subfigure}
  \end{center}
    \floatfoot{\textsc{Notes to Figure \ref{RDDdensity}}:
      Left-hand panel provides a histogram of all Social Protection
      Scores (``Ficha de Protecci\'on Social'') for mothers matched
      to their children's birth records.  The vertical dashed line
      indicates 13,484 points, the cut-off point for Chile Crece Contigo's
      preferential benefits.  This is defined as the top-end of the
      third quintile of vulnerability scores, though these quintiles
      are defined on all recipients of a score in the country, not just
      mothers.  The right-hand panel documents \citet{McCrary2008}'s
      density test around 13,484, documenting the dispersion of
      observations within 1000 points on either side of the cut-off.
    }
\end{figure}

\begin{figure}[htpb!]
  \begin{center}
    \caption{Event Studies}
    \label{fig:event}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.56]{../replication/results/appendix/event_intensity_peso_trimester.eps}
      \caption{Birth Weight}
      \label{event-peso}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.56]{../replication/results/appendix/event_intensity_lbw_trimester.eps}+
      \caption{LBW}
      \label{event-lbw}
    \end{subfigure}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.56]{../replication/results/appendix/event_intensity_premature_trimester.eps}
      \caption{Prematurity}
      \label{event-prem}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.56]{../replication/results/appendix/event_intensity_gestation_trimester.eps}
      \caption{Gestation}
      \label{event-gest}
    \end{subfigure}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.56]{../replication/results/appendix/event_intensity_talla_trimester.eps}
      \caption{Size at Birth}
      \label{event-size}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.56]{../replication/results/appendix/event_intensity_fDeathRate_trimester.eps}
      \caption{Fetal Deaths}
      \label{placebo-fdeaths}
    \end{subfigure}
  \end{center}
  \vspace{-6mm}
  \floatfoot{\textsc{Notes to Figure \ref{fig:event}}: Event studies present estimated models
    interacting ChCC treatment intensity with pre- and post-treatment indicators for each
    3 month period pre- and post-reform.  Here, the ChCC measure refers to average levels
    of ChCC use in the entire post-treatment period (to allow a constant treatment intensity
    for interaction), and this is interacted with indicators for the rollout of the ChCC program
    to each municipality.}
    %The precise specification is: \vspace{-4mm}
    %\begin{equation}
    %InfantHealth_{ct} = \alpha_0 + \sum_{j=-9}^9 \beta_{j} 1\{\text{Time to Adoption} = j\}_t \times \overbar{ChCC}_{c}  + \mu_c + \lambda_t + \varepsilon_{ct}. \vspace{-4mm}
    %\end{equation}  
    %As is standard, 1 period pre-treatment is the omitted reference group.  Periods greater
  %than 9 trimesters pre or post program are indicated in a single $\geq 9$ term.}
\end{figure}


\begin{figure}[htpb!]
  \begin{center}
    \caption{Descriptive RD plot with smaller bins for Social Vulnerability Score (Birth weight)}
    \label{RDD_bw5}
    \includegraphics[scale=1.5]{../replication/results/appendix/cutoffPost_bw5_peso.eps}
  \end{center}
    \floatfoot{\textsc{Notes to Figure \ref{RDD_bw5}}:
      Descriptive plot displays average birth weight outcomes in
      5 point bins of the Social Protection Score, with a separate
      polynomial fitted on each side of the cut-off.  This Figure
      replicates Figure \ref{fig:RD}(a), however now using bins of
      5 points, rather than 55 points, for the running variable.
    }
\end{figure}

\begin{figure}[htpb!]
  \begin{center}
    \caption{Impact of FPS cut-off point on the Probability of ChCC Usage}
    \label{RDDchccplot}
    \includegraphics[scale=1.2]{../replication/results/main/cutoffPost_chcc.eps}
  \end{center}
  \vspace{-4mm}
    \floatfoot{\textsc{Notes to Figure \ref{RDDchccplot}}:
      Descriptive plot documents the probability that mothers are enrolled
      in the ChCC program around the official cut-off for the receipt of
      preferential benefits targeted at the bottom three quintiles of
      recipients of the Social Protection Score.  When estimating a
      regression discontinuity specification in a local linear model
      with \citet{Calonicoetal2014}'s optimal bandwidth, the additional
      likelihood of of participating in ChCC when located just below the
      cut-off is 0.0065(0.019) (coefficient and standard error).
    }
\end{figure}

\begin{figure}[htpb!]
  \begin{center}
    \caption{Variation in Home Visit Intensity by Municipality}
    \label{fig:VDvariation}
    \includegraphics[scale=1.2]{../replication/results/appendix/homeVisits.eps}
  \end{center}
  \vspace{-4mm}
  \floatfoot{\textsc{Notes to Figure \ref{fig:VDvariation}}:
      Histogram documents the average quantity of ``Integral Home Visits''
      received by each targeted family per municipality in Chile in 2013.
      A value of 1 refers to a situation where (on average) each family
      flagged to require a visit based on ChCC's administrative criteria
      receives one visit during the gestational period.  These data are
      averaged for each municipality, and are based on the year 2013 only.
      These data are released by the Ministry of Health (available
      at \url{http://chcc.minsal.cl/indicadores/resultados/293}) and are
      not available for earlier years.  One small municipality with an
      average number of visits of 14.5 per flagged family was removed
      to simplify graphical presentation.}
\end{figure}

%XXX
\begin{figure}[htpb!]
  \begin{center}
    \caption{Health Services and Municipalities}
    \label{fig:healthServices}
    \includegraphics[scale=0.76]{../replication/results/appendix/Servicio_Salud.eps}
  \end{center}
  \vspace{-6mm}
  \floatfoot{\textsc{Notes to Figure \ref{fig:healthServices}}: Municipalities
    are indicated by municipal boundaries, and health services are indicated
    by colours.  Each of Chile's 346 municipalities belongs to one of 29 Health
    Services.  The entire country is displayed at right, and the densely populated
    Metropolitan Region of Santiago is displayed at left.
  }
\end{figure}
\clearpage

\thispagestyle{empty}
\begin{figure}[htpb!]
  \begin{center}
    \caption{ChCC roll-out and Pregnancy Inputs Disbursed}
    \label{mech-plots}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/appendix/ChCCmechanism_controlesPrenatales.eps}
      \caption{Prenatal Check-Ups}%
      \label{mech-prenatal}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/appendix/ChCCmechanism_homeVisits.eps}
      \caption{Home Visits}
      \label{mech-visits}
    \end{subfigure}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/appendix/ChCCmechanism_puritaFortificada.eps}
      \caption{Fortified Milk (Original Formula)}
      \label{mech-purita}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/appendix/ChCCmechanism_puritaMama.eps}
      \caption{Fortified Milk (Updated Formula)}
      \label{mech-mama}
    \end{subfigure}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/appendix/ChCCmechanism_asistenciaSocial.eps}
      \caption{Social Assistance Appointments}
      \label{mech-social}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.6]{../replication/results/appendix/ChCCmechanism_ChileSolidario.eps}
      \caption{Chile Solidario}
      \label{mech-CS}
    \end{subfigure}
  \end{center}
  \floatfoot{\textsc{Notes to Figure \ref{mech-plots}}: Solid blue line
    displays the roll-out of ChCC and proportion of coverage of births
    as in Figure \ref{fig:coverage}.  Dotted red lines display the total
    units of various components of the program disposed over time in
    whole of Chile.  Each panel with the exception of Chile Solidario
    coverage in panel \ref{mech-CS} presents the number of units divided by
    1,000.  Additional discussion of variables and their measurement is
    provided in section \ref{scn:Mechanisms}.
  }
\end{figure}


\clearpage
\setcounter{table}{0}
\renewcommand{\thetable}{D\arabic{table}}
\setcounter{figure}{0}
\renewcommand{\thefigure}{D\arabic{figure}}
\section{Maternal Fixed Effects}
\label{MFE}

%XXX
\begin{table}[htpb!]
  \begin{center}
    \caption{Summary Statistics: Matched Mother, Child and Social Security Data}
    \label{tab:sumstatsMother}
    \begin{tabular}{lccccc} \toprule
      & N& Mean & Std. Dev. & Min & Max \\ \midrule
      \multicolumn{6}{l}{\textbf{Panel A: All Mothers}} \\
      \input{../replication/results/appendix/SummaryMotherAll-update.tex}
      \multicolumn{6}{l}{\textbf{Panel B: Matched Mothers and Children}} \\
      \input{../replication/results/appendix/SummaryMother-update.tex}
      \multicolumn{6}{l}{\textbf{Panel C: Unmatched Mothers and Children}} \\
      \input{../replication/results/appendix/SummaryMotherNoChCC-update.tex}
      \bottomrule
      \multicolumn{6}{p{15.4cm}}{{\footnotesize \textsc{Notes to Table
            \ref{tab:sumstatsMother}}: Summary statistics are presented for all
          births matched with the mother's participation in social programs.
          Summary statistics are presented for all years from 2003-2010.
          \emph{Chile Crece Contigo} began in June of 2007, and so any mothers
          having all births prior to this date never participated in ChCC.  For
          additional notes on variable definitions and comparison with the full
          universe of births (collapsed by municipality) refer to Table
          \ref{tab:sumstats}.}}
      \end{tabular}
  \end{center}
\end{table}


\input{../replication/results/appendix/motherFE.tex}

\input{../replication/results/appendix/motherFE-controls.tex}

\begin{figure}[htpb!]
  \begin{center}
    \caption{Birth weight Distributions Pre- and Post-Program Implementation}
    \label{dists}
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.54]{../replication/results/appendix/Density_weightPre.eps}
      \caption{Birth weights Pre-ChCC}
      \label{predists}
    \end{subfigure}%
    \begin{subfigure}{.5\textwidth}
      \centering
      \includegraphics[scale=0.54]{../replication/results/appendix/Density_weightPost.eps}
      \caption{Birth weights Post-ChCC}
      \label{postdists}
    \end{subfigure} \vspace{-9mm}
  \end{center}
  \floatfoot{\textsc{Notes to Figure \ref{dists}}: Densities are plotted using
    an Epanechnikov kernel with a bandwidth of 5 grams.  Each panel separates
    distributions by whether the mother \emph{ever} participates in Chile Crece
    Contigo.  Panel (a) displays only pre-ChCC time periods, while panel (b)
    displays only post-ChCC time periods.  In both cases, Kolmogorov-Smirnov
    tests reject equality of distributions (in different directions).
  }
  \end{figure}


\clearpage
\renewcommand{\thepage}{A\arabic{page}}




\end{document}
